KataGo 1.12.0 與 UEC 杯用的 model:b18c384nbt-uec.bin.gz

剛剛看到 KataGo 出了 1.12.0,同時也放出了在 2022 年十一月 UEC 比賽時用的 model:「New Neural Net Architecture!」。

1.12.0 比較特別的新的類神經網路架構:

This version of KataGo adds support for a new and improved neural net architecture!

這個新的架構以及其他的改善讓訓練的速度改善:

The new neural nets use a new nested residual bottleneck structure, along with other major improvements in training. They train faster than KataGo's old nets and learn more effectively.

另外一個是他把 UEC 比賽時用的 model 放出來了,很特別的是採用 b18c384,而 KataGo Distributed Training 這邊目前主要是 b40c256 與 b60c320,看起來是為了比賽而一次性訓練出來的。

依照他的說法這個 b18c384 版本跟目前訓練網站上的 b60c320 有差不多強度,但計算速度會比 b60c320 快不少,甚至在一些機器上會跟 b40c256 差不多快:

Attached to this release is a one-off net b18c384nbt-uec.bin.gz that was trained for a tournament in 2022, which should be of similar strength to the 60-block nets on http://katagotraining.org/, but on many machines will run much faster, on some machines between 40-block and 60-block speed, but on some machines even as fast as or faster than 40-block.

另外一個大改變是他把訓練工具從 TensowFlow 跳槽到 PyTorch

The training code has been all rewritten to use pytorch instead of tensorflow.

在 release note 裡沒有提到原因,但這個頗讓人好奇的...